ARCTIC THREE-TOED WOODPECKER 115 



a similar explanation. He says that between 1909 and 1914 there was an irrup- 

 tion of the spruce budwonn in eastern Canada and Maine which resulted in the 

 death of many trees and a consequent increase of bark-beetles and borers, followed 

 by an increase in the number of these woodpeckers. On the other hand a scarcity 

 of the usual food supply may cause migration. A wet season with few fires in 

 the woods or a scarcity of insects (such as the spruce bud-moth) that kill trees 

 might, later, cause a migration. 



Illustrating the length of the sojourn of these woodpeckers during 

 the winter of 1923-24, Dr. Van Tyne (1926) says: 



The greatest concentration of these woodpeckers i*ecorded at any one point 

 was on the estate of Mr. F. H. Keunard where scores of dead and dying white 

 pine afforded an abundance of their special food. The first one seen was a 

 male collected on October 17. Another individual appeared by October 20 and 

 during the winter at least three males and two females were accounted for, 

 while all indications point toward the actual presence of perhaps twice as 

 many. The most remarkable fact about this group of birds, however, was the 

 length of their stay, for both males and females were seen as late as the middle 

 of May and at least one male stayed through the early part of June and was 

 last seen on June 12. 



Other invasions are recorded by Mr. Forbush (1927) as follows: 

 "A great irruption of these birds occurred in the autumn of 1860. 

 During the following winter Mr. George O. Welch often saw as many 

 as six or eight at once in a piece of fire-killed timber in Lynn. 

 * * * In the autumn of 1925, there was a lesser movement, and 

 many returned through New England in the spring of 1926. In the 

 autumn of 1926 another considerable southward migration occurred." 



DISTRIBUTION 



Range. — North America south to the Central United States; non- 

 migratory. 



The range of the Arctic three-toed woodpecker extends north to 

 central Alaska (probably Tocatna Forks and Fairbanks) ; southern 

 Mackenzie (Fort Wrigley, Fort Providence, and Smith Portage) ; 

 northern Manitoba (Cochrane Eiver and probably York Factory) ; 

 Quebec (Kiclmiond Gulf, Godbout, and Madeline River) ; and New- 

 foundland (Nicholsville). East to Newfoundland (Nicholsville) ; 

 probably rarely Prince Edward Island (Baddeck) ; eastern New 

 Brunswick (Tabusintac) ; probably rarely Nova Scotia (Advocate) ; 

 Maine (Machias) ; and probably rarely Massachusetts (Winchen- 

 don and Concord). South to probably rarely Massachusetts (Con- 

 cord) ; central Vermont (Pico Peak) ; southern Ontario (Ottawa, 

 Algonquin Park, and Sand Lake) ; northern Michigan (Au Sable 

 Valley, Blaney, and Huron Mountain) ; probably northern Wiscon- 

 sin (Kelley Brook and Star Lake) ; northern Minnesota (North Pa- 

 cific Junction, Itasca Park, and White Earth) ; probably southwest- 

 ern South Dakota (Elk Mountains) ; northwestern Wyoming (Yel- 



